Recent years have seen great progress in the identification of millions of single nucleotide polymorphisms (SNPs). The abundance of SNPs coupled with the developments of various platforms for high-throughput, high-quality genotyping have made genome-wide association studies within reach for human geneticists. This is further strengthened by the on-going HapMap project that is collecting valuable population genetics information on the allele frequencies and linkage disequilibrium (LD) patterns of millions of SNPs in four different populations, yielding valuable information that can be used to select SNPs in genetic association studies. However, statistical methods lag behind the rapid accumulation of biological information and technology advancement in that existing methods are either not appropriate or not optimized to address scientific questions using these data. Moreover, the new technologies have also led to new questions that are unique and need to be appropriately addressed in a timely fashion to fully realize the potential in these technologies. In this application, we focus on the design and analysis of high-density SNPs in genetic association and linkage studies to identify DMAvariants associated with disease risk. The primary objectives are to investigate study design issues in genome-wide association studies and to develop statistically sound and computationally feasible methods for genetic linkage and association studies. More specifically, four specific aims will be achieved: (1) to investigate strategies in the design of genome-wide association studies; (2) to develop statistical methods for association studies in samples involving individuals having uncertainties on inbreeding patterns and relationships; (3) to develop statistical methods for linkage studies in the presence of LOamong markers; and (4) to implement the developed statistical methods in user-friendly computer programs and make them available to the scientific community.